Artificial Intelligence as a Service,Economic Growth,and Well-Being

Abstract

The share of artificial intelligence (AI) jobs in total job postings has increased from 0.20% to nearly 1% between 2010 and 2019, but there is significant heterogeneity across cities in the United States (US). Using new data on AI job postings across 343 US cities, combined with data on subjective well-being and economic activity, we uncover the central role that service-based cities play to translate the benefits of AI job growth to subjective well-being. We find that cities with higher growth in AI job postings witnessed higher economic growth. The relationship between AI job growth and economic growth is driven by cities that had a higher concentration of modern (or professional) services. AI job growth also leads to an increase in the state of well-being. The transmission channel of AI job growth to increased subjective well-being is explained by the positive relationship between AI jobs and economic growth. These results are consistent with models of structural transformation where technological change leads to improvements in well-being through improvements in economic activity. Our results suggest that AI-driven economic growth, while still in the early days, could also raise overall well-being and social welfare, especially when the pre-existing industrial structure had a higher concentration of modern (or professional) services.

Keywords

Artificial intelligence digital technologies gross domestic product productivity subjective well-being

Introduction

Artificial intelligence (AI) jobs as a share of the total have grown from 0.2% in 2010 to roughly 1% in 2019 with significant potential for continued growth (Perrault et al. 2019). Due to this growth, there are differences in opinion about the consequences of AI on the economy and social welfare (Brynjolfsson and Mcafee 2014; Frey and Osborne 2017). optimists believe the diffusion of AI technology will provide opportunities for economic growth and also improve the state of human well-being. Pessimists share a different view. They believe a small minority of skilled workers (special cities or large corporations) will capture the gains from AI technology leaving other communities more distressed.¹

Like the revolution in digital skills that were observed over the past three decades (Gallipoli and Makridis 2018), the expansion of the demand for AI labor could lead to an expansion of higher-skilled and knowledge-oriented sectors, leading to a new wave of structural transformation (Aghion and Jones, 2018). Although AI jobs are growing at a rapid pace (Perrault et al. 2019), there is significant heterogeneity in AI job growth across cities and sectors. There are also increasing concerns about polarization across cities and the incidence of productivity gains (Autor 2019; Autor and Dorn 2013; Hornbeck and Moretti 2020), at least up until the COVID-19 pandemic.² Due to this, there is growing public attention that high-tech cities like San Francisco and New York take all the gains from growth in the technology revolution, leaving other cities behind (Agrawal, Gans, and Goldfarb 2018; Fernald and Jones 2014; Perrault et al. 2019). But the Covid-19 pandemic has called this phenomenon into question as working-from-home continues to expand, spreading the opportunity for AI jobs beyond high-tech cities (Brueckner, Kahn, and Lin 2021).

In this sense, understanding the relationship between AI, productivity, and subjective well-being (SWB) is important as federal, state, and local policymakers deliberate on whether and how to regulate and support innovation over a potential fourth industrial revolution. Since enlarging the size of the economy and improving well-being are important objectives, this paper integrates the two strands of literature to study the relationship between growth in AI labor demand, economic growth, and well-being. This is because economic growth and well-being are both efficiency and equity consequences of technological progress (Berg and Ostry 2011; Okun and Summers 2015).

After defining AI and laying out a conceptual framework for why we would expect it to affect economic and social activity, we then investigate these claims using empirical data. Our paper contains two important innovations. First, we have both cross-sectional and longitudinal variation obtained by examining AI job growth evolution across varied local labor markets, specifically cities, over time. This allows us to control for national changes and isolate the co-movement of variables at a more granular geographic level. Second, we access data on SWB, which transcends the typical measures of GDP or income per capita. Given the concerns about AI and its potential adverse effects on social welfare, having direct measures of individual well-being offers a more complete view of the distributional consequences of digital transformation. In particular, we provide the first empirical investigation into the role of AI as a leading driver behind transformative human-center services by quantifying the increases in SWB through productivity growth.

The paper provides the first quantitative investigation of the relationship between AI, economic performance, and SWB using new data on AI job postings from Burning Glass Technologies and Gallup’s U.S. Daily Poll across 343 metropolitan areas (Gallup U.S. Poll 2017; Perrault et al. 2019).³ The following observations were made. First, metropolitan areas with greater increases in the share of AI job postings between 2014 and 2018 also exhibit greater economic growth. Second, the positive relationship between AI job growth and economic growth is explained by the initial concentration of technology-based services (or modern service) structure in US cities. Third, and most novel, increases in AI jobs are associated with increases in SWB (especially physical, financial, and social attributes of well-being) where the transmission channel of AI job growth to increased subjective well-being is explained by the positive relationship between AI jobs and economic growth. These results counter the prevailing concern that AI and automation will have net-negative effects on social welfare. Our explicit measurement of SWB, rather than economic performance, is important since standard measures of GDP may overlook unobserved changes in social welfare.⁴

The paper also contributes to the emerging literature on the role of AI in the service economy with broader implications for macro-economic growth and well-being (Huang and Rust 2018; Rust and Huang 2021). Since AI is a service, the paper is also related to the discussion on the implications of AI and its impact on the broader economy (Brynjolfsson, Rock, and Syverson 2019). Although AI can have many positive effects on productivity and the reallocation of tasks (Brynjolfsson et al., 2018; Brynjolfsson & McElheran, 2016), AI and automation can also have adverse effects on employment (Acemoglu and Restrepo 2020). Moreover, if automation displaces tasks at a faster rate than new tasks are created, then the economy can converge to a labor share of zero (Acemoglu and Restrepo 2018a, 2019). As AI becomes capable of performing intuitive and empathetic tasks, some are concerned that it could pose a threat to human employment (Huang and Rust 2018). Nonetheless, as long as prices respond to scarcity and capital is mobile, labor will retain a comparative advantage (CEA 2019a).

Lastly, the paper offers an interdisciplinary and unique data-driven perspective to formalize the links between AI, service revolution, structural transformation, and human well-being (Ghani and Kharas 2010; Herrendorf, Rogerson, and Valentinyi 2014; Mishra, Lundström, and Anand 2011; Mishra, Tewari, and Toosi 2020). In classical economic theory, an economy’s growth potential lies in its innate capabilities that are rooted in its production structure. This can be attributed to the early ideas that viewed growth and development as a process of structural transformation in the productive structure of the economy (Hirschman 1959; Kaldor 1967; Kuznets and Murphy 1966; Lewis 2013), that is, the reallocation of resources from low productivity to high productivity activities. Services are typically viewed as a low-productivity sector and a passive input into the production of physical goods. Yet service activities can be unbundled, disembodied, and splintered in a value chain like goods (Bhagwati 1984). Emerging technologies like automation, artificial intelligence, and big data are changing the characteristics of service. These ideas were originally pioneered at the onset of the modern services revolution as manufacturing began to decline in developed countries, although many questions remain (Karmarkar 2004, 2015; Karmarkar, Kim, and Rhim 2015; Karmarkar and Apte 2007). Furthermore, rapid urbanization and growth in city clusters, characteristic of emerging cities, reduce demand for physical goods and increase demand for services. Agglomeration and economics of scale in service cities are incentivizing service hubs of exports even across US cities (Berube 2007), similar to the factory-driven industrialization in Asian economies.

However, the process of structural change can lead to inequality. Building on empirical literature in economics that uses data on SWB to understand social welfare (Deaton 2012; Kahneman and Deaton 2010), and responding to a call for more research on the relationship between AI and SWB (Musikanski et al. 2020), our paper contributes to the ongoing debate about the effects of AI on social welfare. For example, Amartya Sen remarked that “… quite clearly the notion of living standard and that of utility or happiness are not identical. The former may be typically very important to the latter, but they are not the same thing.” (Sen 1988, 1993). Accounting for these dynamic facets of the economic system, our results quantitatively characterize the linkages between AI jobs, service revolution, economic growth, and well-being.

Definitions and Conceptual Framework

Although we recognize that there is a lot of debate about what AI is and is not, we defer to the definition from the Future of Artificial Intelligence Act of 2017 (H.R. 4625) that AI is “any artificial system that performs tasks under varying and unpredictable circumstances, without significant human oversight, or that can learn from their experience and improve their performance… They may solve tasks requiring human-like perception, cognition, planning, learning, communication, or physical action.” Artificial intelligence often uses machine learning to form predictions and make adaptive adjustments based on new information in their environment (Russell and Peter, 2010). However, it has such a wide array of use cases across sectors that many economists view it as a general-purpose technology (Agrawal, Gans, and Goldfarb 2019). Automation, in contrast, focuses on reducing the labor intensity of a specific process without any stochasticity, whereas AI/ML involves stochasticity and thus, requires human-like decision.

We now lay out the conceptual framework for why we would expect to see a relationship between AI, economic activity, and social activity. On one hand, the expansion of AI could raise productivity growth by increasing the productivity of particularly high value-added tasks that constitute a larger portion of aggregate output than others; such services could include, for example, financial and technology services. On the other hand, the expansion of AI could lead to stagnation in productivity growth if the requisite human capital infrastructure is not built out. That is, if AI technology expands, without a complementary increase in the required human capital there could be short-term productivity gains that do not translate into the medium to long term.

CEA (2019b) argues that in theory, the former mechanism will outweigh, much like the episodes of technological change that we have seen throughout history. In particular, there are three potential phases for the introduction of AI: (a) in the short run, businesses bid wages up as they try to attract high-skilled labor to conducive AI-intensive tasks, (bmaking) in the medium run, AI substitutes for some of the available demand for labor, and (c) in the long run, investment catches up with AI and real wages continue to grow as the talent pool for high-skilled workers grows. Although the story here is theoretically plausible, empirical analysis, especially those related to the modern service economy, has been limited. One prominent study is Brynjolfsson, Rock, and Syverson (2021). To measure the productivity effects of general-purpose technologies, like AI, it is important to account for complementarity between AI and intangible investments, including human capital. In other words, to realize the full productivity effects of AI, certain investments are needed in the earlier years, but the benefits are not borne out in the data until later. Brynjolfsson, Rock, and Syverson (2021) find that productivity is 15.9% higher in 2017, relative to reported statistics. These intangibles are especially present in the modern service economy, which exhibits the greatest demand for AI tasks.

Developed countries have experienced a significant amount of growth in the post-World War II era. Initially driven by a move from agriculture to manufacturing (Herrendorf, Rogerson, and Valentinyi 2014), this technological change led to an increase in the digital workforce across the manufacturing and services sectors (Gallipoli and Makridis 2018). Coupled with the rise of globalization (Rodrik 2018), the rise of digital technologies led to an exodus of workers from manufacturing to services. Note that the modernization of the services sector saved it from dragging down aggregate GDP, at least in the U.S. (Gallipoli and Makridis 2018), where services were traditionally known for having lower productivity growth, that is, “Baumol’s cost disease”(Baumol 1967).

Macroeconomic theory suggests that, if AI and automation lead to a decline in labor intensity and the overall share of manufacturing and agriculture, and less productive sectors of the economy begin to expand, then aggregate GDP could be constrained by what is important and hard to improve and not by what nations are good at (Aghion, Antonin, and Bunel 2019). If the rise of AI and automation is not met with a corresponding increase in skilled labor, polarization in the labor market could rise even further (Autor 2014, 2015; Frank et al. 2019). Using data on labor demand as a proxy, we examine whether disparate growth in AI labor demand is yielding broad-based economic growth across the sample of metropolitan areas or core-based statistical areas (CBSAs). We also provide empirical evidence on the relationship between AI growth and economic growth using the pre-sample share of employment in services to help explain the linkage.

Emerging technologies (e.g., automation, artificial intelligence, and big data) are changing the characteristics of services. Service activities can be unbundled, disembodied, and splintered in a value chain like goods (Bhagwati 1984). Such technological changes are driving up demand for complex technology-intensive services. The virtual capabilities of this class of services, such as being transported cheaply and swiftly in binary bits, make it more desirable than goods.

We call this class of services modern services or professional services, and they include services such as artificial intelligence, information, or research and development, which are different from traditional services or non-professional services or low-tech services, like haircuts and restaurants that require face-to-face contact and more manual labor (Baumol 1967; Ghani and Kharas 2010; Mishra, Lundström, and Anand 2011).⁵ In brief, modern services are digitally intensive and intangible (Gallipoli and Makridis 2018). Similar to high-value-added manufacturing, modern services (e.g., AI) may yield greater knowledge spillovers, have a greater potential for backward and forward linkages, or offer an easier pathway toward discovering new specializations with similar characteristics (Loungani et al. 2017; Spatafora, Anand, and Mishra 2012).

The old idea of services being non-transportable, non-tradable, and non-scalable no longer holds for a range of modern impersonal services (Ghani and Kharas 2010). Over time, AI will be capable of performing intuitive and empathetic tasks, enabling innovative ways for human-machine integration that can impact the largest sector of the global economy—services (Huang and Rust 2018; Rust and Huang 2021). Scholars have extolled the impact of industrialization of services before the advent of the 21^st century with the outsourcing and offshoring only part of the revolution (Bhagwati 1984; Blinder 2006; Karmarkar 2004). The usage of intensive data processing, big data, and AI will become even more pronounced in the coming years as AI-based services have a significant impact on the transformation of many industries, including education (Butler-Adam 2018), retail (Teller, Kotzab, and Grant 2006), human-resources (Gulliford and Parker Dixon 2019), marketing (Rust and Huang 2014), customer engagement (Hollebeek, Sprott, and Brady 2021; Paluch and Wirtz 2020; Singh and Satish Nambisan, 2021), and manufacturing (Li et al. 2017). The pace of AI growth has accelerated over the last decade.

Historically, buyers and sellers needed to meet face to face. Now, many such services can be carried out in one location and consumed in many different places. The internet and other systems of network technologies like industrial robotics, and artificial intelligence are providing technical changes to production techniques and business processes. For example, machine translation systems have significantly increased international trade on various online platforms, increasing exports by 10.9%. This is a result of the substantial reduction in translation costs (Brynjolfsson, Rock, and Syverson 2019). The co-existence of such complex tradable services yields connections to many other complex activities involving both products and services (Mishra, Tewari, and Toosi 2020; Zaccaria et al. 2018). As AI-based services increasingly become the main component of all hardware systems, they begin to possess a physical presence like goods which implies that they can be produced, stored, and consumed across borders (Ghani and O’Connell 2017; Loungani et al. 2017). AI-based services are not only inputs in the global value chain but also standalone services that could impact growth through their growing tradability. The paper uses dis-aggregated CBSA data as an identification mechanism to study AI from the perspective of spatial units within a country. The promise of the service revolution is that location (or economic geography) becomes relatively less important when compared to capital-intensive manufacturing. The characteristics of cities (including industrial structure, skills, and institutions) are important in understanding how the AI-based service revolution could transform into broad-based economic growth and well-being.

The globalization of services is the tip of the iceberg (Blinder 2006; Ghani and Kharas 2010). Services are the largest sector in the world, accounting for more than 70% of global output. The service revolution has altered the characteristics of services. Services can now be produced and exported at a low cost (Bhagwati 1984). Technological changes are transforming intangible services into tangible goods where information technology (IT) serves as a facilitator, enabler, the context, or the service itself blurring the distinction between goods and services (Huang and Rust 2013).

Technological changes are transforming intangible services into tangible goods where information technology (IT) serves as a facilitator, enabler, the context, or the service itself blurring the distinction between goods and services (Brynjolfsson, Collis, and Eggers 2019; Kahneman and Deaton 2010). Given the recent divergence between productivity and median earnings (Stansbury and Summers 2018), some are concerned about the continued effects of technological change on welfare and polarization in the labor market (Korinek and Stiglitz 2019). Even if AI and automation have positive effects on productivity, the effects on welfare and the distribution of welfare could be negative. Introducing well-being as a yardstick for measuring the effects of AI puts people at the center of economic policy questions.⁶ Although there is a general recognition that GDP per capita and SWB are highly correlated (Helliwell, Layard, and Sachs 2012; Stevenson and Wolfers 2008), the correlation can break down, particularly in lower-income countries where there are many other unobserved dis-amenities resulting from weaker institutions and poor healthcare infrastructure.

Recent studies have summarized threats and opportunities facing community well-being for which AI could have positive or negative ramifications (Musikanski et al. 2020). On one hand, there are many potential benefits to particular sectors and communities, including the automotive sector (Li et al. 2018), financial services (Wall 2018; West and Allen 2018), medicine (Topol 2019), education, health, and retail sectors (Breazeal 2019), among others (Musikanski et al. 2020). On the other hand, there are some concerns, which include the devaluation of human skills, declines in self-determination, loss of human control, and displacement of human responsibility (Floridi et al. 2018; Musikanski et al. 2020).⁷ Moreover, there is additional evidence that digital technologies, specifically social media and smartphones have a deleterious impact on youth and the elderly. These deleterious impacts may include lack of sleep, isolation, and depression (Twenge and Park 2019). Artificial intelligence also has the potential to compound issues relating to bias, privacy, data ownership, personal identity, data governance, manipulation, and trustworthiness (Chatila and Havens 2019; IEEE Standards Association 2018).

In addition to productivity—which is reflected in real GDP and per capita income indicators—we consider the effect of AI on SWB. In particular, even if AI leads to greater productivity growth, it could have other unintended consequences on social welfare and the delivery of services, ranging from less social interaction to privacy and security threats. However, we posit that AI in the long run could benefit SWB through its effect on economic activity. For example, if AI leads to productivity growth, as Brynjolfsson, Rock, and Syverson (2021) predict, then it should pass through to wages and per capita income (Hornbeck and Moretti 2020). Our measure of SWB builds upon a large literature that has linked the outlook and attitude of respondents about their lives with income across (Clark, Frijters, and Shields 2008) and within (Deaton 2012; Kahneman and Deaton 2010) countries.

Based on this background, we make the following hypotheses.

H1) As the share of AI jobs increases in a city, productivity increases.

H2) As the share of AI jobs increases in a city, subjective well-being among the average respondent increases.

H3) The increase in subjective well-being resulting from the rise of the AI share of jobs is in part a function of the increase in productivity.

In the next section, we begin by introducing the data and measurement strategy. Then, we take these theoretical hypotheses to the data using a sample of CBSAs that we follow over time.

Data and Measurement

Artificial intelligence requires several disparate inputs to be put into production. These inputs include human capital, software, data, computational power, and management practices. Since firms are still adjusting inputs to AI and experimenting with the technology, current observed investments are only noisy indicators (Tambe 2014). For instance, rapid changes to AI technologies can change the value of investments in AI-related skills (Rock 2019). The challenges in measuring AI apply to both AI inputs and AI outputs as AI is often intangible and changing at a rapid pace. For example, open-source software and data are not transacted through physical markets—so their value is not always fully reflected in the national accounts.⁸ Even beyond quantitative measurement challenges, the scientific understanding of cognitive technologies and the microeconomic processes involved with broader system interactions create barriers to a more complete understanding of AI.

It is important to note that robotics is not necessarily AI. For example, studies show that robot adopters are larger and grow faster than their competitors, experience significant declines in labor share and the share of production workers in employment, and exhibit increases in value-added and productivity (Acemoglu et al., 2020). Studies on robotic automation show that automation has a stronger impact on employment in developing regions. For example, the impact of robotization on employment in emerging economies is more than ten times those of advanced economies. The impact comes from both a direct effect of automation on low-skilled jobs and a reshoring effect (Barbieri et al., 2021). Additionally, robot per thousand workers leads to a reduction of the employment-to-population ratio by 0.18–0.34 percentage points and wages by 0.25–0.50 percentage points between 1990 and 2007 (Acemoglu and Restrepo 2020). However, by studying a cross-panel of countries over the same period, Graetz and Michaels (2018) find that the expansion of robotics led to an increase in labor productivity growth, but only a decline in the employment share for low-skilled workers. Industrial robots increase labor productivity, total factor productivity, and wages. Although they do not significantly change total hours worked, they may be a threat to low- and middle-skilled workers. Economic studies of AI should be explicit about defining AI, as many robotic statistics may not reflect AI. Indeed, many industrial robot shipments have very little (or no) AI in them, which makes it a poor metric for progress in AI (Shoham et al. 2018). To address these measurement challenges, researchers are using AI labor demand from job postings (Mishra, Tewari, and Toosi 2020; Perrault et al. 2019) and skill taxonomies to generate measures of investments in skills related to data and algorithmic decision-making since job postings map to labor resources (Brynjolfsson et al. 2020; Tambe 2014; Tambe, Ye, and Cappelli 2020).

We measure AI from the universe of online job postings scraped by Burning Glass Technologies, spanning over 45,000 online job sites and representing over 85% of the total labor demand (Burning Glass Technologies 2019). Our sample is less likely to cover industries and firms that search for recruits through physical means, like the placement of a “help wanted” sign. To identify AI jobs, we search the text of the job posting for at least one AI-related skill.⁹ The overall measurement strategy is fairly robust to the exact taxonomy; see Acemoglu et al. (2020); Alekseeva et al. (2021) as examples of studies that have used this taxonomy. Generally, data on job postings have been used a lot within labor economics to study the demand for skills across space and time (Hershbein and Kahn 2018). We focus on CBSAs as our primary geography of analysis since they best reflect consolidated labor markets.

The growth in AI labor demand exhibits two notable trends across sectors and regions. According to AI Index (Perrault et al. 2019), the United States' share of AI jobs grew from 0.3% in 2012 to 0.8% of total jobs posted in 2019. Across sectors, AI labor demand is growing especially in high-tech services and the manufacturing sector. Across regions, evidence points to no clear convergence in AI labor demand growth, that is, many small metropolitan areas with a low initial stock of AI jobs also experienced fast growth in AI labor demand (2010–19). The growth of AI labor demand in smaller cities and regions of the US illustrates the potential of AI to generate new types of work across sectors and regions. Figure 1 shows the percent of AI jobs posted as a share of total online jobs posted in 2018–19 across the top 100 metropolitan areas. Artificial intelligence jobs are relatively concentrated in certain pockets of the US. The chart presents a yellow–green color spectrum where yellow indicates the share of AI jobs as a percent of total jobs in the district is below 0.01%, and the green indicates the share of AI jobs as a percent of total jobs is over 1%.

Figure 1.

The heterogeneity in AI job postings across the USA, 2018–2019. Source: Burning Glass Technologies, 2020.

The data on AI job postings are combined with two additional sources. First, traditional measures of economic performance from the Bureau of Economic Analysis (BEA), including per capita income, employment, and GDP. We deflate per capita income by the 2012 personal consumption expenditure index and GDP by the 2012 implicit price deflator from the St. Louis Federal Reserve. Second, dimensions of well-being and human flourishing from Gallup, the United States' premier polling service. We specifically leverage their U.S. Daily Poll, which conducted daily surveys of roughly 1000 U.S. adults on well-being between 2008 and 2018.¹⁰ Gallup’s current polling relies on live interviews (roughly 200 interviewers) with dual-frame sampling (including random-digit-dial [RDD]) landline and wireless phone sampling with randomly sampled respondents (age 18 or over) from all 50 states and the District of Columbia. The sampling method also uses a three-call design to reach respondents who do not pick up on the original attempt.

We focus on the respondents who receive the “well-being track,” which captures five components of well-being: career, community, physical, financial, and social. We enumerate each of the inputs into these overall indices in Section A.1 of the Online Appendix, but we briefly summarize them here. Career well-being focuses on how people enjoy their work, community focuses on their satisfaction with their city of residence, physical focuses on their health and medical conditions, financial focuses on their ability to pay bills and purchase what they need, and social focuses on their friendships and optimism. Although these are subjective, and individuals are not always well-informed about their options (Benjamin et al. 2012), these measures are reliable (Krueger and Schkade 2008) and have been used to study the relationship between income and happiness within (Kahneman and Deaton 2010) and across countries (Stevenson and Wolfers 2008).

Figure 2 builds upon these past studies by plotting a binned scatterplot across the major CBSAs. Although the relationship between logged real per capita income and well-being is positive, it is far from perfect (only 0.15), suggesting that SWB likely detects meaningful variation that is not captured by standard macroeconomic indicators, such as income or GDP. Such data on SWB are especially important when considering the effects of AI and automation because there could also be increases in average productivity (e.g., Graetz and Michaels (2018)), if these technologies are primary complements to capital and benefit capital holders more than laborers. To our knowledge, we are not aware of existing work that has related SWB with AI exposure at the micro-level.

Figure 2.

Subjective well-being and real per capita income, 2014–2018. Source: Bureau of Economic Analysis, Gallup, 2014–2018. The figure plots the association between standardized (mean zero, standard deviation of one) overall well-being and logged real per capita income. Section A.1 of the Online Appendix describes the specific questions used in these indices in more detail.

Table 1 presents summary statistics of our main variables of interest. We report the mean and standard deviations of these variables separately for the full sample, the CBSAs with high-tech services, and the CBSAs with lower-tech services. High-tech refers to whether the CBSA has an employment share in professional services, information services, or finance, insurance, and real estate (FIRE) that is above the median. Low-tech refers to whether the CBSA has an employment share in wholesale trade, retail trade, education, and healthcare above the median. Starting with the AI share, we see that the AI share of IT jobs is 3.1%. Interestingly, CBSAs with more low-tech services have a slightly higher share of IT jobs. However, the AI share of total jobs is higher in the CBSAs with high-tech services, but only 0.16% in the aggregate. This follows from the fact that many IT jobs reside in the low-tech services economy, which require fewer cognitive skills (e.g., call center jobs). We also report our measures of SWB, demographics, industry composition of the cities, and economic activity.

Table 1.

Summary statistics on AI growth, CBSAs, and demographics.

	Full sample		High-tech services		Low-tech services
	Mean	SD	Mean	SD	Mean	SD
AI intensity
AI, % IT	0.031	0.033	0.029	0.021	0.034	0.040
AI, % total	0.0016	0.0022	0.0023	0.0029	0.0010	0.0012
AI, % IT change	0.006	0.029	0.007	0.015	0.005	0.037
AI, % total change	0.000	0.001	0.001	0.001	0.000	0.001
Subjective well-being
Overall SWB	62	2	62	2	61	2
Career SWB	61	2	60	2	61	3
Community SWB	62	4	62	3	61	4
Physical SWB	60	2	61	2	60	3
Financial SWB	61	4	61	3	61	4
Social SWB	61	2	61	2	61	3
Demographics
Population	628,823	1,122,251	1,137,733	1,520,760	239,817	326,495
Male	0.49	0.01	0.49	0.01	0.49	0.01
Age under 18, %	0.24	0.03	0.24	0.03	0.24	0.03
Age 17–34, %	0.24	0.05	0.24	0.04	0.24	0.05
Age 35–54, %	0.28	0.02	0.28	0.02	0.27	0.02
Age 55–64, %	0.11	0.01	0.11	0.01	0.11	0.02
Age 65+, %	0.13	0.03	0.13	0.03	0.13	0.03
White, %	0.80	0.12	0.78	0.11	0.81	0.12
Black, %	0.11	0.11	0.12	0.11	0.10	0.11
Married, %	0.51	0.04	0.51	0.04	0.51	0.04
Some college, %	0.30	0.30	0.30	0.04	0.30	0.04
College, %	0.16	0.04	0.19	0.04	0.14	0.03
Masters, %	0.07	0.02	0.08	0.02	0.06	0.02
Professional, %	0.02	0.01	0.02	0.01	0.01	0.00
Doctorate, %	0.01	0.01	0.01	0.01	0.01	0.01
Industry composition
Information services, %	0.02	0.01	0.02	0.01	0.02	0.00
FIRE, %	0.06	0.02	0.07	0.02	0.05	0.01
Professional services, %	0.09	0.02	0.11	0.02	0.07	0.01
Wholesale/Retail trade, %	0.15	0.02	0.15	0.01	0.15	0.02
Education/Health, %	0.23	0.04	0.23	0.04	0.24	0.05
Economic activity
Real GDP, millions 2012 USD	43	110	84	158	11	13
Income/Capital, 2012 USD	45,211	7162	48,082	7152	43,018	6351
Real GDP growth	0.02	0.03	0.02	0.02	0.01	0.03
Income/Capita growth	0.02	0.03	0.03	0.02	0.02	0.03
Observations	1318		571		747
Height

Notes. Source: Burning Glass Technologies (BGT), Bureau of Economic Analysis (BEA), Census Bureau, 2010–2018. The table reports the means and standard deviations of the main variables of interest: AI job intensity, subjective well-being (SWB), demographics over the city, industry composition of the city, and economic activity (deflated in 2012 terms). High-tech refers to whether the CBSA has an employment share in professional services, information services, and FIRE that is above the median; low-tech refers to whether the CBSA has an employment share in wholesale trade, retail trade, education, and healthcare above the median.

Armed with data on the intensity of AI jobs at a local level, the results identify the links between AI job growth, the service economy, economic growth, and well-being. Figure 3 documents the correlation between various employment shares in different industries at a CBSA level between 2014 and 2018 and the growth in AI jobs as a share of total jobs between 2011 and 2019 for 343 CBSAs. It is not surprising that we see a strong negative correlation (p = −0.21) between the employment share in agriculture, mining, and forestry and the employment share in AI. This is consistent with the large macroeconomic literature on the transition of developed countries from agriculture (Herrendorf, Rogerson, and Valentinyi 2014). Although we see a null association with the manufacturing share, the correlation with low-skilled services (retail and wholesale trade) is strongly negative (p = −0.43), and the correlation with high-skilled services (information, professional, and finance, real estate, and insurance) is strongly positive (p = 0.60), consistent with the rise in digital workers and the changing composition of the labor market (Gallipoli and Makridis 2018).

Figure 3.

Employment composition and the 2011–2019 growth in the share of AI jobs. Source: Burning Glass Technologies (BGT) and Census Bureau. The figure plots the correlation (population-weighted) between the share of workers in the (i) agriculture, mining, and forestry sector; (ii) the manufacturing sector; (iii) the low-tech (wholesale and retail trade) services sector; and (iv) the high-tech (finance, professional, and information) services sector and the 2011 to 2019 percentage point change in the share of artificial intelligence jobs in a core business statistical area (CBSA).

The knowledge and technology spillovers of modern services are robust. In the absence of such industrial structure and endowments, AI-based labor demand could remain an isolated enclave, rather than serve as a catalyst for economy-wide growth like we posit in our first hypothesis. In more distorted economies, there may be less scope for inter-sectoral and intra-sectoral resource allocation, as well as knowledge and technology spillovers. To test this hypothesis (the first hypothesis), the formal regression specification is as follows:

y_{u} = γ A l_{u} + f (X_{l}, θ) + ϕ_{l} + λ_{t} + ε_{u}

(1)

where

y_{u}

denotes our measure of productivity (log real GDP and per capita income) in location (CBSA) l and year t,

A l

denotes the AI share of jobs,

f (X_{l} θ)

denotes a semi-parametric function of location-specific characteristics, such as demographics and industry composition,

ϕ

and

λ

denote fixed effects on location and year. Although the location-specific characteristics are collinear with our location-fixed effects, we layer our specification sequentially. Standard errors are clustered at the CBSA level to allow for auto-correlation (Bertrand et al. 2004).

Table 2 documents the results associated with equation (1). Starting with columns 1 and 8, we see that there is a strong positive association between economic growth and the AI share of jobs: a 1% rise in the AI share of jobs is associated with a 25% and 9.5% increase in real GDP and income per capita. Importantly, these estimated coefficients are obtained after controlling for a semi-parametric function of location-specific demographic characteristics, such as the age, education, and race distributions. Columns 2 and 9 subsequently introduce industry composition controls, but the coefficients are fairly invariant. We then add CBSA and year-fixed effects in columns 3 and 10, which control for time-invariant heterogeneity across space and time. Consistent with our concerns about upward bias, the gradients decline in magnitude, but remain statistically and economically meaningful: a 1 percentage point rise in the AI share of jobs is associated with a 7% and 2.7% increase in real GDP and per capita income.

Table 2.

The relationship between the AI share of jobs and productivity.

	log(Real GDP)							log(Real Income/Capita)
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)
AI share of total	25.32∗∗∗	23.80∗∗∗	7.10∗∗∗	8.57∗∗∗	−16.61∗∗∗	8.93∗∗∗	−9.71∗∗	9.50∗∗∗	10.44∗∗∗	2.77∗∗∗	−7.68∗∗∗	3.28∗∗∗	−4.45∗∗∗
	[4.64]	[5.05]	[1.94]	[2.09]	[3.69]	[2.28]	[3.90]	[2.69]	[3.00]	[0.99]	[1.82]	[1.22]	[1.69]
× high manufacturing				−2.64
				[3.89]
× high skilled services					118.20∗∗∗						52.13∗∗∗
					[16.76]						[8.41]
× low skilled services						−6.25∗∗						−1.75
						[2.62]						[1.59]
× high college							50.84∗∗∗						21.85∗∗∗
							[11.06]						[5.01]
R-squared	0.952	0.955	0.999	0.999	0.999	0.999	0.999	0.480	0.535	0.964	0.964	0.964	0.964
Sample size	2565	2565	2563	2563	2563	2563	2563	2565	2565	2563	2563	2563	2563
CBSA controls	Yes	Yes	No	No	No	No	No	Yes	Yes	No	No	No	No
Industry shares	No	Yes	No	No	No	No	No	No	Yes	No	No	No	No
CBSA FE	No	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes
Year FE	No	No	Yes	Yes	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Yes
height

Notes. Source: Burning Glass Technologies (BGT), Bureau of Economic Analysis (BEA), 2010–2018. The table reports the coefficients associated with regressions of logged real GDP (in 2012 prices) and logged per capita income on the share of AI jobs in the CBSA × year. CBSA controls include: logged population, the share male, the age distribution (under age 18, 18–34, 35–54, 55–64, and 65+), race (white and black), the education distribution (some college, college, masters, professional, and PhD). Industry controls include: the share in agriculture, mining, forestry, the share in construction, the share in manufacturing, the share in wholesale and retail trade, the share in information services, the share in finance, real estate, and insurance, the share in professional services, and the share in education and healthcare. High manufacturing, high skilled services (information, professional, and FIRE), low skilled services (retail and wholesale trade, education, and healthcare), and college or more are indicators constructed to equal one if the CBSA share is above the median in the sample of CBSAs in the matched BGT/Census data. Standard errors are clustered at the CBSA-level.

To understand if these associations only reflect the productivity gains that have arisen from the broader digital revolution, that is, those discussed in Gallipoli and Makridis (2018), we parse our data to identify a broader set of digital (information technology) jobs, subsequently running a horse race between the two job shares. When our outcome is logged real GDP, we find a much lower and statistically insignificant association of 0.09 (ρ-value = 0.785). We also find a negative association when our outcome is logged per capita outcome. In sum, these results show that variation in the share of AI jobs captures systematically different movements in the data than the share of IT jobs (the correlation is only 0.54), suggesting that AI-driven innovation in the modern service economy has different and new prospects for the emerging digital economy.

We now turn toward several dimensions of heterogeneity. We create indicators for whether the CBSA has above the median share of workers in the manufacturing sector, the high skilled services sector (information, professional, and finance, real estate, and insurance), low skilled services (retail trade, wholesale trade, education, and healthcare), and college attainment. We do not find evidence of heterogeneous treatment effects across CBSAs that rank higher in their employment share of manufacturing workers, but we do find large and statistically significant effects for CBSAs that rank above the median with respect to differences in the industrial composition of services and educational attainment.

For example, a 1 percentage point rise in the AI share of jobs is associated with a large 118% rise in real GDP and a 52% rise in real income per capita for CBSAs that have above the median employment share of high skilled services. However, there is a slightly smaller increase for those that rank below the median. In contrast, we find that a similar increase in the AI share of jobs is associated with only a slight positive increase in economic activity for CBSAs that rank above the median employment shares of low-skilled services. Finally, we find substantially larger effects of AI in CBSAs that rank above the median in their share of college degree holders. These results are consistent with macroeconomic models of capital-skill complementarity and skill-biased technical change (Katz and Murphy 1992).

The results are related to economic literature that shows how changes in the relative prices between sectors can lead to a shift in industrial composition (Ngai and Pissarides 2007). If AI is more of a complement to higher-skilled and digitally intensive services sectors (Gallipoli and Makridis 2018), then the expansion of AI jobs could lead to an increase in their share of the economy. Moreover, increases in the skill premium (Buera, Kaboski, and Rogerson 2015), particularly for AI and technology workers, could also lead to changes in industrial composition.

Figure 4 plots the raw data across CBSAs grouped using a binned scatterplot. Panel A shows that there is a correlation of 0.22 between the 2014 and 2018 increase in the share of AI jobs and the growth of overall well-being and Panel B shows that there is a correlation of 0.51 between the 2010 and 2017 increase in the share of AI jobs and the growth in per capita income. Admittedly, the correlation could be confounded by a number of unobserved forces, but we find the raw correlations motivating and useful as a starting point. The regressions that follow in the subsequent paragraphs endeavor to control for potentially spurious factors and explore whether these results remain robust.

Figure 4.

Changes in the AI share of jobs, well-being, and income. Source: Bureau of Economic Analysis, (Perrault et al. 2019), Gallup, 2010–2018. Panel A plots the 2014–2017 growth in CBSA well-being with the percentage point change in the AI share of jobs across CBSAs weighted by the number of respondents in the CBSA. Panel B plots the 2010-2017 growth in CBSA per capita income with the percentage point change in the AI share of jobs across CBSAs weighted by employment.

We use a different regression model to understand the effects of AI on well-being as a broader measure of welfare. Also, this model controls for economic activity to assess whether the impact of AI on well-being is coming through economic activity. The formal regression specification is as follows

S W B_{u} = γ A l_{u} + ζ y_{u} + ϕ_{l} + λ_{t} + ε_{u}

(2)

where

S W B

denotes our measure of standardized subjective well-being and

y

denotes our measures of productivity as additional control variables.

Table 3 documents the results associated with equation (2). Although we find a statistically significant positive association between increases in the AI share of jobs and well-being across every specification without the productivity-based controls (odd columns), these correlations decline for all except physical well-being. In particular, a 1 percentage point rise in the AI share of jobs is associated with a 34SD (standard deviation) rise in physical well-being. Since the average AI share of jobs is 0.00,186, then the association comes to 0.063SD when evaluated at the mean. The fact that the correlations decline in statistical significance after adding the productivity-related controls suggests that increases in the AI share of jobs affect well-being primarily through productivity effects.

Table 3.

The relationship between the AI share of jobs and subjective well-being.

	Overall		Career		Community		Physical		Financial		Social
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)
AI share of total	44.89∗∗	10.36	24.23	27.72	42.30∗∗	5.70	31.31∗	34.66∗	59.94∗∗∗	9.82	34.99∗∗	−18.61
	[18.68]	[17.34]	[20.75]	[19.75]	[17.69]	[19.54]	[17.39]	[18.63]	[20.24]	[18.08]	[17.54]	[18.23]
log(Real GDP)		1.30		0.52		1.33		0.82		0.54		0.47
		[0.97]		[1.08]		[0.82]		[0.83]		[1.04]		[1.22]
log(Real Income/Capita)		1.40		1.68∗		-.34		1.67∗∗		2.43∗∗		0.53
		[0.95]		[1.00]		[1.42]		[0.85]		[1.21]		[1.17]
R-squared	0.55	0.63	40	0.48	0.70	0.75	0.55	0.63	0.48	0.55	0.42	0.45
Sample size	1666	1307	1666	1307	1666	1307	1666	1307	1666	1307	1666	1307
CBSA FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Height

Notes. Source: Burning Glass Technologies (BGT), Gallup, Bureau of Economic Analysis (BEA), 2010–2018. The table reports the coefficients associated with regressions of standardized z-scores of subjective well-being, conditional on CBSA and year fixed effects. Subjective well-being is measured in five ways: career, community, purpose, financial, and social, which combine (unweighted average) into overall well-being. Section A.1 of the Online Appendix describes the specific questions used in these indices in more detail.

To more precisely understand the mediating role of productivity on subjective well-being, we use methods from Tingley et al. (2014) who introduced an approach for causal mediation analysis. In brief, we find a marginal effect of 46.35, which is statistically indistinguishable from our main result in the manuscript of 44.59 in column 1 of Table 3. This approach comes with the added benefit of allowing us to identify how much of the overall effect is driven by productivity gains, like real GDP. We find that the proportion mediated is 77.6% (ρ-value = 0.062). This suggests that the majority of the differences we see in subjective well-being are driven by productivity effects, but not all, which offers an even more optimistic message.

Even with granular controls and fixed effects, there are still other confounding effects we may not be addressing. The correlation between AI and SWB declines in statistical significance after we control for economic characteristics, suggesting that economic performance mediates the benefits of AI on SWB. Nonetheless, it is hard to distinguish between all the underlying mechanisms at play. For example, there could be knowledge spillovers that stem from universities and other innovative environments (Moretti 2004), leading to the effective deployment and application of AI in a community. Moreover, there could be complementarities between certain industrial compositions and specific AI-related skills (Autor 2015). Furthermore, demographic factors, especially age distribution, might influence the returns to human capital and the acquisition of AI skills (Acemoglu and Restrepo 2018b).

In unreported regressions, we also explore the relationship between the share of high-tech services and well-being in the cross-section (i.e., without CBSA fixed effects). We find a strong positive correlation of 0.27. The correlation remains strong and statistically significant until the share of college graduates and/or professional degree holders is controlled for, which makes the association statistically insignificant at conventional levels. This is consistent with the view that modern services are associated with greater well-being, but they alone are not a causal factor. Indeed, only when they are coupled with the introduction of technology do they advance well-being and productivity, as we have shown above.

Discussion and Conclusion

This paper presents the first empirical evidence on the relationship between AI jobs, the service revolution, economic growth, and subjective well-being (SWB). The results highlight the central role that service-based cities play to translate the benefits of AI job growth to subjective well-being. Cities with greater growth in AI job postings also exhibit greater economic growth. The relationship between AI job growth and economic growth is driven by cities that had a higher concentration of modern (or professional) services. Artificial intelligence job growth also leads to an increase in the state of well-being in US cities. The transmission channel of AI job growth to increased subjective well-being is explained by the positive relationship between AI jobs and economic growth. These results are consistent with models of structural transformation where technological change leads to improvements in well-being through improvements in economic activity. Our results counter the prevailing concern that AI and automation will have net-negative effects on social welfare, and suggest AI-driven economic growth could have a positive effect on well-being—concentrated in cities with a greater initial industrial structure of modern services.

Policymakers should take these results as a source of optimism about the effect of AI on not only the economy but also people. Attempts to curb or over-regulate AI activities may adversely affect both economic growth and consumer welfare. Instead, policymakers should continue establishing frameworks for AI so that a market for “AI as a service” can emerge and help organizations and people make better and more effective decisions. A good example involves the principles of trustworthy AI established in 2020, which articulates a set of ethical guidelines for the development and expansion of AI.¹¹

One limitation of these results, however, is that we are still in the infancy of AI—that is, the vast majority of jobs still do not require AI skills. There could be a non-linear relationship between economic & social outcomes and the AI share. For example, when AI job shares are low, the economy and society boom since it is an important complement to digital services, but as the AI share grows, it can create greater inequality and polarization in the labor market. Although we recognize the possibility that higher levels of AI jobs could have different effects than the ones we document, we emphasize the role of continued human capital accumulation in the workforce. As long as individuals are upskilling, that should shield them from displacement. In conclusion, regions with higher shares of AI job growth are likely to reap the economic and social benefits that come with it. Therefore, local and national policymakers should leverage endowments in their service capacity to maximize the positive influence of technological changes to help shape the well-being of citizens for long-term prosperity.

Nonetheless, some measurement issues still need to be investigated further to understand the effects of AI on the economy and society. First, AI is a diverse technology and contains many fundamental measurement challenges in identifying and creating a community-accepted definition of skills/technologies to classify what is an AI job and what is not. Second, we only consider the channel of AI labor demand in this study. There are other potential channels to measure AI growth, including publications, patents, papers, and private, or public investment. Third, future research could include attributes of income/wealth distribution besides well-being. However, data and measurement on all these attributes are difficult, especially at a local level. We leave these for future research.

Supplemental Material

Supplemental Material - Artificial Intelligence as a Service, Economic Growth, and Well-Being

Supplemental Material for Artificial Intelligence as a Service, Economic Growth, and Well-Being by Christos A. Makridis and Saurabh Mishra in AI Service and Emotion

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Saurabh Mishra

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Acemoglu

Daron

Autor

David

Hazell

Jonathon

Pascual

Restrepo

, Andreas Mueller, Rob Seamans and Betsey Stevenson (2020), “AI and Jobs: Evidence from Online Vacancies,” Cambridge, MA: NBER Working Paper w28257. DOI: 10.3386/W28257.

Acemoglu

Daron

Pascual

Restrepo

(2018a), “The race between man and machine: Implications of technology for growth, factor shares, and employment,” American Economic Review. DOI: 10.1257/aer.20160696.

Acemoglu

Daron

Pascual

Restrepo

(2018b), “Demographics and Automation,” NBER working paper. Oxford Academic. DOI: 10.1093/RESTUD/RDAB031.

Acemoglu

Daron

Pascual

Restrepo

(2019), “Automation and new tasks: How technology displaces and reinstates labor, ” Journal of Economic Perspectives. 33, 3-30, DOI: 10.1257/jep.33.2.3.

Acemoglu

Daron

Pascual

Restrepo

(2020), “Robots and jobs: Evidence from us labor markets, ” Journal of Political Economy, 128. 2188-2244, DOI: 10.1086/705716.

Aghion

Jones

, (2018), CI Jones-The economics of artificial intelligence: An and undefined 2018. “Artificial intelligence and economic growth,” nber.org.

Aghion

Philippe

Antonin

Céline

Simon

Bunel

(2019), “Artificial intelligence, growth and employment: The role of policy,” Economie et Statistique, 2019. DOI: 10.24187/ecostat.2019.510t.1994.

Agrawal

Ajay

Gans

Joshua

Goldfarb

Avi

(2018), “Prediction machines: the simple economics of artificial intelligence,” Harvard Business Press, 250.

Agrawal

Ajay

Gans

Joshua

Goldfarb

Avi

(2019), “Economic policy for artificial intelligence,” Innovation Policy and the Economy, University of Chicago Press, 19, 139-159. DOI: 10.1086/699935.

10.

Alekseeva

Azar

Gine

Samila

, Taska-Labour economics and undefined 2021 (2021), “The demand for AI skills in the labor market,” Elsevier.

11.

Autor

David H.

(2014), “Skills, education, and the rise of earnings inequality among the ‘other 99 percent, Science. 344, 843-851. DOI: 10.1126/science.1251868.

12.

Autor

David H.

(2015), “Why are there still so many jobs? the history and future of workplace automation,” Journal of Economic Perspectives. 29, 3-30. DOI: 10.1257/jep.29.3.3.

13.

Autor

David H.

(2019), “Work of the Past, Work of the Future, ” AEA Papers and Proceedings, 109, 1-32, DOI: 10.1257/pandp.20191110.

14.

Autor

David H.

Dorn

David

(2013), “The Growth of Low Skill Service Jobs and the Polarization of the U.S. Labor Market, ” American Economic Review, 103, 1553-1597.

15.

Bainbridge

Lisanne

(1983), “Ironies of automation,” Automatica, 19, 3. DOI: 10.1016/0005-1098(83)90046-8.

16.

Barbieri

Laura

Mussida

Chiara

Piva

Mariacristina

Vivarelli

Marco

(2021), “Testing the Employment Impact of Automation, Robots and AI: A Survey and Some Methodological Issues, ” SSRN Electronic Journal. DOI: 10.2139/ssrn.3457656.

17.

Baumol

William J.

(1967), “Macroeconomics Of Unbalanced Growth: The Anatomy Of Urban Crisis,” The American Economic Review, 57, 10.

18.

Benjamin

Daniel J.

Heffetz

Ori

Miles

, (2012), Kimball and Alex Rees-Jones (2012), “What do you think would make you happier? What do you think you would choose?, ” American Economic Review, 102. 2083-2110, DOI: 10.1257/aer.102.5.2083.

19.

Berg

Andrew G.

Ostry

Jonathan D.

(2011), “Equality and efficiency,” Finance and Development.

20.

Bertrand

Marianne

Duflo

Esther

Mullainathan

Sendhil

Thank

Alberto

Abadie

Acemoglu

Daron

Angrist

Joshua

Banerjee

Abhijit

Chernozhukov

Victor

Hirano

Kei

Guido

Imbens

Katz

Larry

Rey Kling

Lang

Kevin

Levitt

Steve

Murphy

Kevin

Saez

Emmanuel

Staiger

Doug

Topel

Bob

Nber

Cepr

Nber

(2004), “How much should we trust differences-in-differences estimates?,” Quarterly Journal of Economics, 119, 249-275.

21.

Berube

Alan

(2007), MetroNation: How US metropolitan areas fuel American prosperity, Metropolitan Policy Program at Brookings.

22.

Bhagwati

Jagdish N.

(1984), “Splintering and Disembodiment of Services and Developing Nations,” World Economy, 7, 17. DOI: 10.1111/j.1467-9701.1984.tb00265.x.

23.

Blinder

Alan S.

(2006), “Offshoring: The next industrial revolution?, ” Foreign Affairs. 85, 113, DOI: 10.2307/20031915.

24.

Breazeal

Cynthia

(2019), “Living and flourishing with AI,” vimeo. com/313938302 .

25.

Brueckner

Jan

Kahn

Matthew

Lin

Gary

(2021), “A New Spatial Hedonic Equilibrium in the Emerging Work-from-Home Economy?” NBER Working Paper.

26.

Brynjolfsson

Erik

Collis

Avinash

Eggers

Felix

(2019), “Using massive online choice experiments to measure changes in well-being,” Proceedings of the National Academy of Sciences of the United States of America, 116, 27. DOI: 10.1073/pnas.1815663116.

27.

Brynjolfsson

Erik

Mcafee

Andrew

(2014), “The digitization of just about everything,” The second machine age: work, progress, and prosperity in a time of brilliant technologies. Goodreads.

28.

Brynjolfsson

Erik

McElheran

Kristina

(2016), “Data in Action: Data-Driven Decision Making in U.S. Manufacturing, ” SSRN Electronic Journal. 56. DOI: 10.2139/ssrn.2722502.

29.

Brynjolfsson

Erik

Mitchell

Tom

Rock

Daniel

(2018), “What Can Machines Learn and What Does It Mean for Occupations and the Economy?, ” AEA Papers and Proceedings, 108. 43-47, DOI: 10.1257/pandp.20181019.

30.

Brynjolfsson

Erik

Rock

Daniel

Syverson

Chad

(2021), “The Productivity J-Curve: How Intangibles Complement General Purpose Technologies, ” American Economic Journal: Macroeconomics, 13. 333-372, DOI: 10.1257/mac.20180386.

31.

Brynjolfsson

Erik

Tambe

Prasanna

Hitt

Lorin M.

Rock

Daniel

(2020), “IT, Intangible Capital , and Superstar Firms,” NBER. NBER Working Paper .

32.

Buera

Francisco J.

Kaboski

Joseph P.

Rogerson

Richard

(2015), “Skill-Biased Structural Change,” NBER. NBER Working Paper . DOI: 10.1093/restud/rdab035.

33.

Burning Glass Technologies (2019), “Mapping the Genome of Jobs: The Burning Glass skills taxonomy,” Burning Glass Technologies.

34.

Butler-Adam

John

(2018), “The Fourth Industrial Revolution and education, ” South African Journal of Science. 114, DOI: 10.17159/sajs.2018/a0271.

35.

CEA (2019a), Adapting to Technological Change with Artificial Intelligence while Mitigating Cyber Threats, Washington, DC: CEA.

36.

CEA (2019b), Economic Report of the President, Washington, DC: CEA.

37.

Chatila

Raja

Havens

John C.

(2019), “The IEEE global initiative on ethics of autonomous and intelligent systems,” In Intelligent Systems, Control and Automation: Science and Engineering. DOI: 10.1007/978-3-030-12524-0_2.

38.

Clark

Andrew E.

Paul

Frijters

Shields

Michael A.

(2008), “Relative income, happiness, and utility: An explanation for the Easterlin paradox and other puzzles, ” Journal of Economic Literature, 46. 95-144, DOI: 10.1257/jel.46.1.95.

39.

Corrado

Carol A.

Hulten

Charles R.

(2010), “How do you measure a ‘technological revolution’?” In American Economic Review. DOI: 10.1257/aer.100.2.99.

40.

Corrado

Carol

Haltiwanger

John

Daniel Sichel (2005), Measuring Capital in the New Economy, Measuring Capital in the New Economy, University of Chicago Press. DOI: 10.7208/chicago/9780226116174.001.0001.

41.

Corrado

Carol

Hulten

Charles

Daniel Sichel (2009), “Intangible capital and U.S. economic growth,” Review of Income and Wealth. Wiley, DOI: 10.1111/j.1475-4991.2009.00343.x.

42.

Deaton

Angus

(2012), “The financial crisis and the well-being of Americans, ” Oxford Economic Papers, Oxford Economic Papers, 64. 1-26, DOI: 10.1093/oep/gpr051.

43.

Fernald

John G.

Jones

Charles I.

(2014), “The future of US economic growth,” In American Economic Review. Wiley, DOI: 10.1257/aer.104.5.44.

44.

Floridi

Luciano

Cowls

Josh

Beltrametti

Monica

Chatila

Raja

Chazerand

Patrice

Dignum

Virginia

Luetge

Christoph

Madelin

Robert

Ugo

Pagallo

Rossi

Francesca

Schafer

Burkhard

Valcke

Peggy

Vayena

Effy

(2018), “AI4PeopleAn Ethical Framework for a Good AI Society: Opportunities, Risks, Principles, and Recommendations,” Minds and Machines, 28, 1, DOI: 10.1007/s11023-018-9482-5.

45.

Frank

Morgan R.

Autor

David

Bessen

James E.

Brynjolfsson

Erik

Cebrian

Manuel

Deming

David J.

Feldman

Maryann

Groh

Matthew

Lobo

José

Moro

Esteban

Wang

Dashun

Youn

Hyejin

Rahwan

Iyad

(2019),“Toward understanding the impact of artificial intelligence on labor,” Proceedings of the National Academy of Sciences of the United States of America. DOI: 10.1073/pnas.1900949116.

46.

Frey

Carl Benedikt

Osborne

Michael A.

(2017), “The future of employment: How susceptible are jobs to computerisation? ” Technological Forecasting and Social Change, 114. 254-280, DOI: 10.1016/j.techfore.2016.08.019.

47.

Gallipoli

Giovanni

Makridis

Christos A.

(2018), “Structural transformation and the rise of information technology, ” Journal of Monetary Economics, 97. 91-110, DOI: 10.1016/j.jmoneco.2018.05.005.

48.

Gallup U.S. Poll (2017), The U.S. Daily Poll, 2008-2017. Gallup U.S. Poll.

49.

Ghani

Ejaz

Kharas

Homi

(2010), “The Service Revolution in South Asia: an Overview,” The Service Revolution in South Asia. South Asia Region.

50.

Ghani

Ejaz

O’Connell

Stephen D.

(2017), “Can services be a growth escalator in low-income countries? ”, OKR. 24.

51.

Graetz

Georg

Michaels

Guy

(2018), “The Review of Economics and Statistics ROBOTS AT WORK,” MIT Press, 100.

52.

Gulliford

Fred

Dixon

Amy Parker

(2019), “AI: the HR revolution,” Strategic HR Review, 18. DOI: 10.1108/shr-12-2018-0104.

53.

Helliwell

Layard

Sachs

(2012), “World happiness report.”

54.

Herrendorf

Berthold

Rogerson

Richard

Valentinyi

Ákos

(2014), “Growth and Structural Transformation,” In Handbook of Economic Growth. DOI: 10.1016/B978-0-444-53540-5.00006-9.

55.

Hershbein

Brad

Kahn

Lisa B.

(2018), “Do recessions accelerate routine-biased technological change? evidence from vacancy postings,” American Economic Review, 108. 1737-1772, DOI: 10.1257/aer.20161570.

56.

Hirschman

I. I.

(1959), “On multiplier transformations, ” Duke Mathematical Journal, 26. DOI: 10.1215/S0012-7094-59-02623-7.

57.

Hollebeek

Linda D.

Sprott

David E.

Brady

Michael K.

(2021), “Rise of the Machines? Customer Engagement in Automated Service Interactions, ” Journal of Service Research. 24, 3-8, DOI: 10.1177/1094670520975110.

58.

Hornbeck

Richard

Moretti

Enrico

(2020), The Growth of Low Skill Service Jobs and the Polarization of the U.S. Labor Market.

59.

Huang

Ming Hui

Rust

Roland T.

(2013), “IT-Related Service: A Multidisciplinary Perspective, ” Journal of Service Research, 16. 251-258, DOI: 10.1177/1094670513481853.

60.

Huang

Ming Hui

Rust

Roland T.

(2018), “Artificial Intelligence in Service, ” Journal of Service Research, 21. 155-172, DOI: 10.1177/1094670517752459.

61.

IEEE Standards Association (2018), “The IEEE global initiative on ethics of autonomous and intelligent systems,” IEEE Standards Association. https:/1standards.ieee.org/developlindconn/ec/autonomous_systems

62.

Inman

Robert P.

(1989), Managing the Service Economy : Prospects and Problems, Cambridge Books.

63.

Jones

Charles I.

Klenow

Peter J.

(2016), “Beyond GDP? Welfare across countries and time, ” American Economic Review, 106. 2426-2457, DOI: 10.1257/aer.20110236.

64.

Kahneman

Daniel

Deaton

Angus

(2010), “High income improves evaluation of life but not emotional well-being,” Proceedings of the National Academy of Sciences of the United States of America, 107. 16489. DOI: 10.1073/pnas.1011492107.

65.

Kaldor

(1967), “Strategic factors in economic development.”The University of Chicago Press.

66.

Karmarkar

Uday

(2004), “Will you survive the services revolution? ” Harvard Business Review, 82, 100-107.

67.

Karmarkar

Uday

(2015), “OM Forum- The service and information economy: Research opportunities,” Manufacturing and Service Operations Management. 17, 20. DOI: 10.1287/msom.2015.0525.

68.

Karmarkar

Uday S.

Apte

Uday M.

(2007), “Operations management in the information economy: Information products, processes, and chains, ” Journal of Operations Management, 25. 438-453, DOI: 10.1016/j.jom.2006.11.001.

69.

Karmarkar

Uday S.

Kim

Kihoon

Rhim

Hosun

(2015), “Industrialization, Productivity and the Shift to Services and Information, ” Production and Operations Management, 24. 1675-1695, DOI: 10.1111/poms.12379.

70.

Katz

Lawrence F.

Murphy

Kevin M.

(1992), “Changes in relative wages, 1963-1987: Supply and demand factors, ” Quarterly Journal of Economics, 107. 35-78, DOI: 10.2307/2118323.

71.

Korinek

Anton

Stiglitz

Joseph E.

(2019), “AI and its implications for income distribution and unemployment,” University of Chicago Press.

72.

Krueger

Alan B.

Schkade

David A.

(2008), “The reliability of subjective well-being measures, ” Journal of Public Economics, 92. 1833-1845, DOI: 10.1016/j.jpubeco.2007.12.015.

73.

Kuznets

Murphy

(1966), Modern economic growth: Rate, structure, and spread, New Haven, CT: Yale University Press.

74.

Lewis

W. Arthur

(2013), The theory of economic growth, The Theory of Economic Growth. MIT Press. DOI: 10.4324/9780203709665.

75.

Bo hu

Hou

Bao cun

Wen tao

Xiao bing

Yang

Chun wei

(2017), “Applications of artificial intelligence in intelligent manufacturing: a review,” Frontiers of Information Technology and Electronic Engineering. 18, 86-96. DOI: 10.1631/FITEE.1601885.

76.

Jun

Cheng

Hong

Guo

Hongliang

Qiu

Shaobo

(2018), “Correction to: Survey on Artificial Intelligence for Vehicles, ” Automotive Innovation, 1. 390, DOI: 10.1007/s42154-018-0034-8.

77.

Loungani

Mishra

Papageorgiou

Wang

(2017), World trade in services: Evidence from a new dataset, International Monetary Fund.

78.

Mishra

Lundström

Anand

(2011), “Service export sophistication and economic growth,” World Bank Policy Research Working Paper.

79.

Mishra

Saurabh

Tewari

Ishani

Toosi

Siavash

(2020), “Economic complexity and the globalization of services,” Structural Change and Economic Dynamics, 53. DOI: 10.1016/j.strueco.2020.03.002.

80.

Moretti

Enrico

(2004), “Workers’ education, spillovers, and productivity: Evidence from plant-level production functions,” American Economic Review. DOI: 10.1257/0002828041464623.

81.

Musikanski

Laura

Rakova

Bogdana

James

Bradbury

Phillips

Rhonda

Manson

Margaret

(2020), “Artificial Intelligence and Community Well-being: A Proposal for an Emerging Area of Research, ” International Journal of Community Well-Being, 3. 39-55, DOI: 10.1007/s42413-019-00054-6.

82.

Ngai

L. Rachel

Pissarides

Christopher A.

(2007), “Structural change in a multisector model of growth, ” American Economic Review, 97. 429-443, DOI: 10.1257/aer.97.1.429.

83.

Okun

Arthur M.

Summers

Lawrence

(2015), Equality and efficiency: The big tradeoff. DOI: 10.2307/2148425.

84.

Paluch

Stefanie

Wirtz

Jochen

(2020), “Artificial Intelligence and Robots in the Service Encounter, ” Journal of Service Management Research, 4. 3-8, DOI: 10.15358/2511-8676-2020-1-3.

85.

Perrault

Raymond

Yoav Shoham Brynjolfsson

Erik

Clark

Jack

Etchemendy

John

Grosz

Barbara

Lyons

Terah

James

Manyika

Mishra

Saurabh

Carlos Niebles

Juan

(2019), “Artificial Intelligence Index 2019 Annual Report,” Human-Centered AI Institute, Stanford University.

86.

Rock

Daniel

(2019), “Engineering Value: The Returns to Technological Talent and Investments in Artificial Intelligence, ” SSRN Electronic Journal. DOI: 10.2139/ssrn.3427412.

87.

Rodrik

Dani

(2018), New technologies, global value chains, and developing economies. NBER.

88.

Russell

Stuart

Peter

Norvig

(2010), Artificial intelligence a modern approach, London: Pearson Education Inc.

89.

Rust

Roland T.

Huang

Ming Hui

(2014), “The service revolution and the transformation of marketing science, ” Marketing Science, 33. 206-221, DOI: 10.1287/mksc.2013.0836.

90.

Rust

Roland T.

Huang

Ming-Hui

(2021), The Feeling Economy, The Feeling Economy, Springer International Publishing. DOI: 10.1007/978-3-030-52977-2.

91.

Sen

Amartya

(1988), “Chapter 1 The concept of development,” In Handbook of Development Economics. Elsevier. DOI: 10.1016/S1573-4471(88)01004-6.

92.

Sen

Amartya

(1993), “Capability and well-being73, ” The Quality of Life, 30, 1–445.

93.

Shoham

Yoav

Perrault

Raymond

Brynjolfsson

Erik

Clark

Jack

James

Manyika

Carlos Niebles

Juan

Lyons

Terah

Etchemendy

John

Grosz

Barbara

Bauer

Zoe

(2018), The AI index 2018 annual report. Stanford University.

94.

Jagdip

Singh

Satish Nambisan

2021, Gary Bridge and Jürgen Kai Uwe Brock (2021), “One-Voice Strategy for Customer Engagement, ” Journal of Service Research, 24. 42-65, DOI: 10.1177/1094670520910267.

95.

Smith

Aaron

Anderson

Monica

(2017), “Americans’ attitudes toward a future in which robots and computers can do many human jobs,” Pew Research Center.

96.

Spatafora

Anand

Mishra

(2012), Structural transformation and the sophistication of production. IMF.

97.

Stansbury

Anna

Summers

Lawrence H.

(2018), “Productivity and Pay: Is the Link Broken?, ” SSRN Electronic Journal. . DOI: 10.2139/ssrn.3192609.

98.

Stevenson

Betsey

Wolfers

Justin

(2008), “Economic growth and subjective well-being: Reassessing the Easterlin paradox,” In Brookings Papers on Economic Activity. NBER. DOI: 10.2139/ssrn.1273524.

99.

Tambe

Prasanna

(2014), “Big data investment, skills, and firm value, ” Management Science, 60. 1452-1469, DOI: 10.1287/mnsc.2014.1899.

100.

Tambe

Prasanna

Xuan

Cappelli

Peter

(2020), “Paying to program? Engineering brand and high-tech wages, ” Management Science, 66. 3010-3028, DOI: 10.1287/mnsc.2019.3343.

101.

Teller

Christoph

Herbert

Kotzab

Grant

David B.

(2006), “The consumer direct services revolution in grocery retailing: An exploratory investigation, ” Managing Service Quality, 16. 78-96, DOI: 10.1108/09604520610639973.

102.

Tingley

Dustin

Yamamoto

Teppei

Hirose

Kentaro

Keele

Luke

Imai

Kosuke

(2014), “Mediation: R package for causal mediation analysis, ” Journal of Statistical Software, 59. DOI: 10.18637/jss.v059.i05.

103.

Topol

Eric J

. (2019), “High-performance medicine: the convergence of human and artificial intelligence, ” Nature Medicine. 25, 44-56, DOI: 10.1038/s41591-018-0300-7.

104.

Twenge

Jean M.

Park

Heejung

(2019), “The Decline in Adult Activities Among U.S. Adolescents, 1976–2016,” In Child Development, Blackwell Publishing Inc., 90, 638–654. DOI: 10.1111/CDEV.12930.

105.

Wall

Larry D.

(2018), “Some financial regulatory implications of artificial intelligence, ” Journal of Economics and Business, 100. 55-63, DOI: 10.1016/j.jeconbus.2018.05.003.

106.

West

Darrell

Allen

John

(2018), “How artificial intelligence is transforming the world,” UK: cemas.org.

107.

Zaccaria

Andrea

Mishra

Saurabh

Cader

Masud

Pietronero

Luciano

(2018), Integrating Services in the Economic Fitness Approach. DOI: 10.1596/1813-9450-8485.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB